# Forecasting People's Needs in Hurricane Events from Social Network

arXiv: https://arxiv.org/abs/1811.04577 · HTML: https://mlsanity.com/paper/1811.04577/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Nov 2018.

## Built on by (most cited first, 4 of 4)

- [A multi-modal approach towards mining social media data during natural disasters -- a case study of Hurricane Irma](https://mlsanity.com/paper/2101.00480/) (2021)
- [Situational-Aware Multi-Graph Convolutional Recurrent Network (SA-MGCRN) for Travel Demand Forecasting During Wildfires](https://mlsanity.com/paper/2304.06233/) (2023)
- [Public Health in Disaster: Emotional Health and Life Incidents Extraction during Hurricane Harvey](https://mlsanity.com/paper/2408.11133/) (2024)
- [Discovering Transmission Dynamics of COVID-19 in China](https://mlsanity.com/paper/2512.22787/) (2025)

## Abstract (arXiv)

Social networks can serve as a valuable communication channel for calls for help, offering assistance, and coordinating rescue activities in disaster. Social networks such as Twitter allow users to continuously update relevant information, which is especially useful during a crisis, where the rapidly changing conditions make it crucial to be able to access accurate information promptly. Social media helps those directly affected to inform others of conditions on the ground in real time and thus enables rescue workers to coordinate their efforts more effectively, better meeting the survivors' need. This paper presents a new sequence to sequence based framework for forecasting people's needs during disasters using social media and weather data. It consists of two Long Short-Term Memory (LSTM) models, one of which encodes input sequences of weather information and the other plays as a conditional decoder that decodes the encoded vector and forecasts the survivors' needs. Case studies utilizing data collected during Hurricane Sandy in 2012, Hurricane Harvey and Hurricane Irma in 2017 were analyzed and the results compared with those obtained using a statistical language model n-gram and an LSTM generative model. Our proposed sequence to sequence method forecast people's needs more successfully than either of the other models. This new approach shows great promise for enhancing disaster management activities such as evacuation planning and commodity flow management.